Hyun Shin’s presentation, at this week’s BoE/IMF/Hong Kong Monetary Authority conference on monetary policy and macro pru, went something like this.
One dimension along which the dollar has become the global currency is in denominating debt, especially in emerging market economies, presumably because of a history of institutional and other local risks there. This happens, increasingly, between parties neither of which are domiciled in the US. At the point where the Fed begins to tighten, any resulting increase in the dollar may set off a debt-deleveraging event. Higher value dollar debt relative to the home currency of the borrower leaves that borrower with a higher debt-service burden, whose per period home currency revenues or income are now worth less.
This makes the borrower more likely to go under, which increases the riskiness of the portfolio of the lender, who may respond by switching assets away into something else; the effect of which may be to aggravate the problem even further, as it starves those same foreign currency borrowers looking to roll over funding, reducing their net worth and increasing their riskiness still further, and so on.
Hyun put this forward as one reason why the Fed ought to have thought carefully about loosening so much in response to the financial crisis. The effect of which, by depressing the dollar relative to the counterfactual, was to encourage dollar-denominated borrowing. Presumably, since we start from here, not 2009, the implication also is that the Fed, having erred by over-doing it, should tighten cautiously [pointed out at the conference by Steve Cecchetti]. Note that these concerns are urged on the Fed even if they are focused solely on their US mandate, because of the effect of a blow-back on the US from these important markets.
In his discussion of Shin, Charlie Bean questioned whether the mechanism he had identified was quantitatively important enough to worry about.
In the same spirit, imagine this concern put before the FOMC as they contemplated the grand loosening strategy in the aftermath of Lehman’s demise. One might charactarise Shin as having articulated a risk, relative to the New Keynesian, finance-free model that the Fed staff use for forecasting, of over-doing the stimulus.
However, recall that at that time central banks were probably estimating that the zero bound was going to deprive them of something like 6-8 percentage points of interest rate stimulus. And they did not have the evidence they comfort themselves with now of the encouraging impact effect on long yields of quantitative easing. Moreover, they would have doubted the willingness of Congress to pass a sufficiently large fiscal stimulus package, and wondered too about its efficacy.
So by far the predominant concern was that there would be far too little stimulus, and the economy would get trapped as the zero bound, as Japan appeared to be, and as the same models with our without finance explained could happen. Indeed, as things have panned out, with a protracted period of below target inflation, and time spent at the zero bound far longer than most predicted, these concerns were arguably proven well founded.
Shin’s risk of stoking a foreign currency-borrowing related spree in emerging markets would, I suggest, have been dwarfed by this concern on the other side. In the absence of any constraints on policy, one might have expected the Fed to fine tune it somewhat to insure themselves against the risks Shin identifies. However, forced to chart a path a long way from the ideal by the many constraints placed upon them, such things would have made no difference at all.